Really.

If, on the one hand, it's ecomonically inefficient for the defendants to have PDs, because they get worse deals for their clients when the charges are "minor", then the Rethugs' policy argument is:  "do away with the requirement for PDs in 'minor' cases, because the defendants get a better deal without them.  That "minor case" exception is the one which will swallow the whole.

Similarly, prosecutors (and cops) can tell defendants "you don't really need a lawyer.  Here's this scientific study and it tells you you'll get a better deal with us than you will through that skanky PD.  So, deal with us."

Moreover, if you look at the criminal justice system through the "administrative" rather than "trial" model - a distinction in which "administrative" minimizes the "jury trial" and "beyond a reasonable doubt" requirements and which Prof. Doug Berman over at Sentencing Law and Policy has explored over the years in the contexts of Blakely/Booker, sentencing-for-uncharged/unconvicted-crimes, and elsewhere - why have PD.  Economically, it reduces the system costs (dollars, cents and time) to not have PDs.  

And, let's not forget the distinction in ineffective assistance cases, where denial of the defendant's private counsel of choice results in a new trial, while denial of the defendant's public defender of choice is harmless error at best.  Scalia made a big show talking about wanting the best lawyer money can buy, during recent oral argument in a case on that distinction, but the distinction remains.

And, eliminating PDs would, of course, aid the administration of the Courts by acting as a "laxative on the docket" (h/t Grits for Breakfast, I think, for the phrase).  This surcease would not last long, though, as idle government employees would have their time sucked up by more cases, not fewer.

Of course, this all comes down to the ultimate (what some would call) amorality of economics - it recognizes only costs and profits, and decrees the former must be eliminated and the latter maximized, and now.  If a dollar sign cannot be placed before it, it doesn't count.  And, sadly, one cannot place a dollar sign before liberty and all those other, quaint concepts written in the Constitution and given lip service by the Rethugs.

As you suggest, it seems particularly strange that they would exclude probation, and only somewhat less so to exclude acquittals.

The study excluded people who received probation.

The study also excluded acquittals at trial and considered only those who received prison sentences.

are PD's that much more inept than private counsel? maybe they get more guilty indigent clients?

Combine that with the questionable decision to account for probation/acquittal simply as zero as well as some other questionable assumptions they made and the fact that the study as written makes it difficult to decipher how/why they addressed variables as they did, and I don't know if we're looking at statistical noise or something meaningful.

Regardless, while the "marginal indigence" theory has at least some intuitive appeal, it seems quite a stretch to suggest that this is a major factor in the discrepancy.  It seems worthy of mention as a possible contributor, but they give it surprising (and I think unsubstantiated) weight in the study as well as the op-ed.  

The authors note, "It is important to recognize that all measures of effectiveness--even sentence outcomes--may have built-in biases that cut against public defenders, including the possibility that public defender clients may tend to be in custody rather than on bond (putting substantially more pressure on them to plea bargain) and that public defender clients may tend to have more prior felonies (subjecting them to greater penalties if convicted). Although our statistical techniques could in theory control for these variables, the data available to us did not allow us to do so."(227)

These two factors, prior criminal history and whether the defendant is in or out of custody, are probably the two most important factors in explaining the different outcomes that result from the same charges (aside from the facts underlying the charges).  

The fact that the researchers were unable to control for custody creates a structural bias within the study against PD's.  PD clients are, by definition, poor and unlikely to post bail/bond.  A defendant's incarceration acts as an institutional pressure upon the defendant themselves to seek a speedy disposition, often in-spite of a more favorable outcome at a later point in time.  From the in-custody defedant's point of view, it is a better choice to end the incarceration ASAP than to remain incarcerated and fight their case.  Since PD's represent a higher percentage of defendants who are incarcerated, the PD disposition statistics will reflect this institutional effect without controlling for it.

The proof of such studies is independent replication. The problem is that such studies are expensive and the independent replication may differ from the original study in important ways.

  Anectdotally, I would say that by far the worst representation comes from  certain privately retained lawyers. Incompetent or insufferably lazy folks do not long survive as PDs because their work is subject to monitoring and supervision. Any hack with a license can claim to be a criminal defense lawyer and criminal defendants are not always the most sophisticated and discerning consumers.

   That said, it is very difficult for PD offices to retain their best because of the nature of the work and compensation. One has to be truly committed to the cause to spend years and years in a PD's office. The disparity in starting salaries while large is something many will bear for a few years to gain invaluable experience,  but down the road people often have far more need for money due to raising a family and the disparity in compensation only grows with experience.

   I think there are more cases of outright ineffective assistance (in real terms not in Strickland terms) with private counsel but, I do also believe that, with a significant number of exceptions, the most talented and experienced folks are found in private practice.

  In sum, I think PDs mostly do a very good job and relatively rarely a inexcusably bad job, but I think the private bar has more people at the extremes --good and bad.